This disclosure relates generally to acoustic signal level detection and control, and more particularly, to hearing safety protection in communication devices that limit the level of an acoustic signal that is transmitted to a listener through a communication device.
Proper control of electrical acoustic signal levels within a communication system is desirable to ensure high quality communication and hearing safety to individuals using the system. Generally, a telephone handset or headset includes a microphone and a speaker. For example, a telephone headset provides a microphone on an arm that is positioned proximate to a user's mouth and a speaker disposed within an earpiece that is positioned over the user's ear. In order to provide proper communication, the sound level of the audible signal emitted from the speaker should fall within a range of sound intensity. If the sound level is below this intensity range, then the user may not be able to hear or understand what is being said by the remote talker. Conversely, if the sound level is above this intensity range, then the user may be subjected to increasingly uncomfortable sound levels finally reaching the point of injury. Loud audible signals are of concern in communication devices, such as telephone handsets and headsets, that position a speaker near a user's ear.
These loud audible signals may be caused by a variety of different events. For example, accidental disturbances within a communication connection may cause an electrical signal level to increase dramatically (e.g., a malfunctioning amplifier, intense feedback, incorrect signal source, or phone line shorted to power line). Oftentimes, the transient time of a signal to reach a high level may be very short and a listener may have little or no time to remove an earpiece or other listening device from her ear before she is exposed to that high level. In the case of a telephone headset user, the listener may have to bring her hand to her ear, which may require a second or two before the earpiece is removed. Also, many headset users spend a large amount of time on the phone. For example, headset users, such as telemarketers, receptionists, and operators often spend eight hours a day answering telephone calls and speaking with people over a telephone. Consequently, these individuals are at a higher risk of sound exposure caused by a sudden or constant loud audible signal. Thus, due to the extended time headset users are on telephone calls and the extra time required to remove the headset, headset users are particularly vulnerable to exposure to loud audible signals generated from the headset speaker.
FIG. 1 depicts a varistor 100 that has been used to reduce the maximum level to the speaker, and hence, the sound exposure. Basically, a varistor 100 is a surge protector that senses peaks within a signal and, in response, creates a shunt path, using diodes, that attenuates these peaks. As shown in FIG. 1, a voltage source 105, with corresponding source impedance 110, provides an electrical signal to an attached load 125 via the varistor 100. Examples of this voltage source 105 include a telephone or telephone adapter, and examples of the load include a telephone handset or headset speaker. The varistor 100 has a first diode 115 connected in parallel to a second diode 120, each having a corresponding turn-on voltage. This turn-on voltage activates the diode, and typically, is between 0.5-0.7 volts. The diodes 115, 120 are used to shunt the positive and negative halves of an electrical signal through their corresponding paths after the electrical signal received from the voltage source 105 reaches the respective diode turn-on voltages.
The first diode 115 is used to shunt the negative half of the electrical signal current and is activated by a magnitude of the negative half exceeding the first diode's turn-on voltage. If this magnitude is below the turn-on voltage, then the first diode 115 remains cut off. However, if the magnitude is above the turn-on voltage, then the first diode 115 is on and conducts current from the voltage source 105, thereby creating a shunt path. The amount of current actually conducted through the shunt path depends on the impedance of the load 125. In any event, the level of the negative half of the electrical signal current received by the load 125 is reduced because of this activated negative shunt path.
The second diode 120 operates in a similar manner as the first diode 115, except that it creates a shunt path for the positive half of the electrical signal received from the voltage source 105. As a result, the magnitude of each half of the electrical signal is attenuated after the diodes 115 and 120 are turned on. However, as is apparent, the voltage across the load 125 is not totally clamped if the electrical signal power increases. Rather, voltage across the load 125 may continue to increase, although at a much lesser rate due to the clamping effects of the positive and negative shunt paths. The voltage increase across the load 125 is caused by the fact that the resistance levels of the diodes 115, 120 are finally fixed. The level of shunting provided by the diodes 115, 120 is dependent on the ratio of the diode resistance and the load resistance, both of which are finally fixed, resulting in the diodes being incapable of adjusting the level of shunting provided by the diode paths. Thus, the voltage across the load 125 may continue to rise and produce a very loud audible signal if the level of the electrical signal is sufficiently high.
The problems above are caused primarily by the non-linear effects of the two diodes 115, 120. Specifically, the relationship between the current through a diode and the voltage across the diode is nonlinear. As the voltage source reaches the turn on voltage of the two diodes 115, 120, the diodes 115, 120 are activated and shunt current from the load 125. Additionally, the nonlinear characteristics of the activated diodes 115, 120 cause higher distortion across the load 125. Thus, the quality of any audible signal generated by the load 125 is compromised as turn on voltage is approached.
FIG. 2 depicts another circuit that has been used to reduce the electrical signal peaks. As shown in FIG. 2, a discrete transistor circuit 200 is coupled in parallel to the first and second diodes 115, 120. The discrete transistor circuit 200 comprises a first transistor 220, a second transistor 225, a first resistor 215 and a second resistor 230. The first transistor 220 base and emitter are connected across voltage source 105. The second transistor 225 is coupled in parallel to the voltage source 105 in a similar manner. This discrete transistor circuit 200 is placed in front of the two diodes 115, 120 and is activated by an electrical signal voltage level from the voltage source 105 being above the turn-on voltage of the first and second transistors 220, 225.
The first transistor 220 attenuates the positive half of an electrical signal after the voltage level between the first transistor 220 emitter and base exceeds the first transistor's turn-on voltage. Typically, this transistor turn-on voltage is between 0.5-0.7 volts. Once this first transistor 220 is turned on, an attenuation network is created, comprising the first resistor 220 and the resistance (Rce1) between the emitter and the collector of the first transistor 220. This attenuation network functions to decrease the positive voltage at the load 125 by allowing current to flow through the first transistor 220. As the voltage level of the electrical signal increases, the transistor 220 goes into saturation mode, and the resistance (Rce1) between the emitter and collector of the first transistor 220 decreases. As a result, because this resistance (Rce1) forms a divider network with the first resistor 215, the current flowing through the first transistor 220 increases, thereby limiting the relative voltage across the load 125. As the voltage source increases, the saturation level within the first transistor 215 deepens. Hence, the voltage across the load 125 decreases.
The second transistor 225 operates in a similar manner as the first transistor 220, except that it operates on the negative half of the electrical signal from the voltage source 105. Specifically, after the second transistor 225 is turned on, an attenuation network is created comprising the second resistor 230 and the resistance (Rce2) between the collector and emitter of the second transistor 225. Thus, the negative half of the electrical signal is attenuated, resulting in limiting the relative voltage across the load 125 as the voltage level on the electrical signal from the voltage source 105 increases.
Although the use of the two transistors 220, 225 provides better attenuation characteristics than the varistor 100 shown in FIG. 1, the two transistors 220, 225 are not capable of sufficiently attenuating a very large signal. This failure results from the saturation characteristics of the two transistors 220, 225. Specifically, if the voltage on the electrical signal increases to a sufficiently high level to drive the transistors 220, 225 into a deep saturation mode, the resistance (Rce2) between the emitter and collector of the transistors 220, 225 becomes fixed. As a result, the two transistors 220, 225 function similar to diodes that operate within the varistor 100 and a resistor network is created. Thus, as the voltage level on an electrical signal increases sufficiently to drive the transistors 220, 225 into deep saturation, the voltage across the load 125 increases.
The first diode 115 and the second diode 120 are still required to provide the Zener short circuit to prevent the output from increasing. In order to compensate for the above-described problem, a varistor 100 may be placed behind the discrete transistor circuit 200. However, the above-described problems regarding varistors are then reintroduced. The resistor network resulting from the diodes 115, 120 is thus not able to clamp the voltage across the load 125, but rather, only to decrease the slope of any voltage increase of the load 125.
FIG. 3 graphically depicts the inability of the above-described prior devices to effectively limit the maximum voltage level across the load 125. As shown in FIG. 3, an electrical output voltage signal 305 operating below a turn-on voltage 320 is relatively unaffected by the discrete transistor circuit 200. However, after the turn-on voltage is reached, the rate at which the electrical output voltage signal 310 increases is reduced. If only the varistor 100 is used, then the voltage across the load 125 will continue to rise 325 as the voltage source 105 increases. If both the discrete transistor circuit 200 and the varistor 100 are used, then the voltage across the load 125 will initially decline 330. However, once the voltage source 105 is sufficiently high to drive the transistors 220, 225 into deep saturation, then the voltage across the load 125 will start to rise 335. Due to the fixed resistance of the deep saturation characteristics of the transistors 220, 225, the voltage across the load 125 is not absolutely clamped, but instead, increases slowly in relation to an increase in the source voltage 105. It is important to note that the output voltage 310 may still rise above a desired clamping level 315.
Accordingly, there is a need for methods and apparatus that limit an output voltage as the voltage source increases. Additionally, because the apparatus may be required to operate in small devices with a limited power, such as in communication headsets, there is an additional need that the apparatus minimize cost, power consumption, and distortion.